Apart from non-melanoma skin cancer, prostate cancer is the most commonly diagnosed cancer among men in the United States today. In spite of its importance, few modifiable predictors of this disease have been established. Well-established risk factors for prostate cancer (e.g., age) are not amenable to modification and, hence, have limited utility as targets of primary prevention strategies. There is a need to identify other risk factors that can be controlled. 5-alpha reductase, the enzyme that converts testosterone to dihydrotestosterone, has been proposed to play a key role in the etiology of prostate cancer; thus, the NCI is currently testing finasteride, a 5-alpha reductase inhibitor, in the prevention of prostate cancer in an ongoing, randomized clinical trial. No information is available on other agents capable of inhibiting 5-alpha reductase and prostate cancer risk in men. Saw palmetto, derived from the berry of the American palm tree, can inhibit 5-alpha reductase activity. Saw palmetto has been shown to be as effective as finasteride in the treatment of urinary symptoms from benign prostatic hyperplasia. No data are available regarding its association with prostate cancer incidence in men; however, in vitro studies show inhibition of growth of prostate cancer cell lines. Therefore, we propose to conduct a retrospective cohort study among 16,700 physicians (mean age 67 years) to test the hypothesis that saw palmetto use reduces the risk of developing prostate cancer. These men currently are being followed as part of two other funded studies, the Physicians' Health Study I and II. Information on the development of prostate cancer is being collected in these two studies, as is a whole host of other information on health habits and medical history. In this application, we propose to retrospectively collect details on the use of saw palmetto (when started and stopped, dose, brand) by adding questions to the scheduled 72-month follow-up questionnaire in these two other studies. During the proposed period of study, we anticipate that 790 cases of prostate cancer will develop. Power calculations show adequate power to detect a 45% reduction in risk of prostate cancer associated with saw palmetto use. At little additional cost, the data from this study will contribute to the knowledge infrastructure regarding whether large scale, informative trials on saw palmetto use and prostate cancer risk should be supported in the future.
